Assessment Flashcards
Introduction
When we select someone for a job, we are trying to predict if that person will be able to perform tasks to a certain standard. This would also be applicable for promotions.
Assessing people correctly is important. Whether they will behave safely – but also prevent stress at work by helping to ensure that people are suited to the jobs they do.
Criterion Related validity refers to the strength of the relationship between a predictor (assessment method) and some criterion of work performance.
Instead of own CRV studies – rely on validity coefficients to similar groups (similar jobs, elsewhere). This is done where samples are not large enough to carry out a validity study – cannot account situational variables such as stress which differ from one work environment to another.
Assessment Centres (AC)
Seek to account for ‘all the variance’ by using multiple methods of assessment with multiple assessors over several days with group of candidates
Reliability and validity dependent on those of assessment methods used and combination thereafter.
Observation: Possibility of the Hawthorne effect,
• Intra reliability: how consistent the analyst is
• Inter-rater reliability: where more than one observer is used to check if scores are consistent as otherwise interpretation could be subjective.
Structured Questionnaires: Standardised Procedure increases reliability and once tested and proven, selection method will always be reliable.
Functional Job Analysis (Fine and Wiley, 1974) uses analyses of constituent task in terms of action sequences and comes up with a percentage of the job directed towards ‘data’, people and other aspects – aids observation (direct, video, etc).
Work Sample Tests:
Any assessment which involves the assesses being asked to perform a behaviour similar to those performed in the work role.
• Psychomotor
• Job Knowledge Tests
• Individual Decision Making
• Group Discussions
• Indoor exercises with leader – Observed and Rated.
If designed appropriately, based on job and person analyses, reliability and validity good.
Establishing performance Criteria: may come from supervisor rating of performance, simple output measures, after recruitment has taken place.
Nature of Interviews:
Interviewer: aim to gather, evaluate and select or reject, who must combine and weigh data to come to a decision. This determines the validity of the interview.
Interpersonal Communication: Interviewee is influenced by the behaviour of the interviewer – paraphrasing, agreeing, disagreement and silence. Behaviour influenced by self-perception of each party, not possible to have standard interviewer – problems with reliability.
Group vs individual interview: validity would not be improved if group decision was made better to have two individual decisions.
Very little predictive power for interviews (Mayfield, 1964; Arvey and Campion 1982).
Studies have shown poor inter-rater reliability: The degree of agreement between two or more interviewers of the same interviewee. Even poor intra-rater reliability. Usual range of biases in social perception when ‘traditional interview’ used.
This could be improved by the 5-point plan:
• Impact on others
• Qualifications
• Experience
• Motivation
• Adjustment
Situational Interviews (SI)
Use job analyses to provide job-related incidents for job-related questions. ‘What
would you do if…?’ For example, the type of questions asked to graduate based on the requirements/ competencies
• Training of interviewers
Other potential selection methods:
Self-assessment, peer assessment, references and educational achievement.
Reliability & Validity of these methods are poor. Low criterion related validity.
Can’t assume the right person will be right for the job. Especially for a graduate scheme job.
